Data Informed Decision Making: A Five-Step Routine for Seeing What the Numbers Miss

Read time —
8 Minutes
Last updated
November 18, 2026
In Short

Data informed decision making is using data as one input into a decision you still make yourself, alongside context, experience and the evidence you go and find. Data-driven decisions let the numbers choose; data informed decisions use them to choose where to look, and keep a person accountable for the call.

In my first months as a field sales executive, I was hitting 75% of target and spending my visits on the wrong customers. What turned it round was data informed decision making, though I'd never heard the phrase. All I had at the time was a regional manager called Tony and a spreadsheet I hadn't looked at properly.

I had more than 300 accounts and 140 visit slots a month. Some accounts were spending, many more weren't, and I was trying to see them all.

"What's the 80/20 split?" Tony asked.

I didn't know. I hadn't even heard of the Pareto principle. A quick sort and one formula later, the spreadsheet showed me that about 60 customers delivered 80% of my sales.

That was the easy part, and it's where most advice about using data stops. The data had told me where to spend most of my time. It couldn't tell me what to do with the other 240.

Those accounts barely bought from us, so on a sales report they all looked much the same. Some were small for good reason. A few were small only because we hadn't understood them, and nothing in the numbers said which was which.

Data is very good at showing you what has already happened, and much weaker at showing you what could. Deciding well with data starts with knowing which of the two you're looking at.

By the end of this article, you'll have a five-step routine for using data on your next decision without handing the decision to the data.

What Is Data Informed Decision Making?

Data informed decision making is using data as one input into a decision you still make yourself, alongside context, experience and the evidence you go and find. The numbers shape the call, and you stay accountable for making it.

Its close relative is data-driven decision making, where the measured evidence chooses between your options. Both start from the same respect for numbers. They part company when the numbers can't see everything that matters.

It also sits next to fact-based decision making, which settles a single disagreement with a checkable fact. Data informed decision making is for the messier calls, where no one fact will settle it and some of what matters hasn't been counted yet.

Data informed means the numbers get a vote, and you still cast the deciding one.

How Is Data Informed Different From Data Driven?

The difference is who has the final word. In a data-driven decision, the data decides and people carry it out. In a data-informed decision, people decide with the data in front of them.

The distinction was popularised by Andrew Chen, now a general partner at Andreessen Horowitz, who says he first heard it at a Facebook talk in 2010. His point was that data is easiest to trust when there's plenty of it, it arrives quickly and it's clear what good looks like. That gives you a quick test for the decision in front of you.

Data-driven Data informed
Who decides The data You, with the data
Works best when Data is plentiful, arrives fast and "good" is clear Data is thin, slow or can't measure what matters
Typical decisions Pricing tests, stock reordering, ad spend Hiring, a key account, a new market, a change of direction
Main risk Optimising what's easy to count Calling it judgement to avoid inconvenient numbers

‍

Most real decisions use both. A sales manager might let the data drive routine reordering and stay data-informed about which new accounts deserve a visit.

If the data is plentiful, fast and clear about what good looks like, let it drive; if any of the three is missing, let it inform.

How Do You Make Data Informed Decisions?

You make data informed decisions by working through five questions, in order, before you commit. Each one stops the data doing a job it isn't fit for, and together they slot into the wider work of good decision-making.

What Decision Are You Actually Making?

Start by naming the decision before you open any data. Write the choice down in one sentence, along with what would make you choose differently.

Open the dashboard first and the dashboard sets the agenda. You end up answering whatever the reports happen to measure, which is rarely the question in front of you.

In my case the decision was easy to state and hard to make: where should 140 visits a month go?

Whose Experience Do Your Numbers Miss?

Next, ask what your data can't see. Every report is built from what someone chose to record, and that's usually whatever is easy to count from the inside.

Sales figures record who bought. They don't record who could have bought, why someone stopped, or what a customer is buying from someone else.

That blind spot has a name, bounded awareness: the information that would change your call often exists, but your frame has drawn the edges somewhere else. Before you decide, list the people who live with the result and ask whether your numbers reflect what they experience.

What Can the Data Tell You, and What Can't It?

Data is strong on what is happening and weak on why. It can show you that an account buys a handful of lines, but it can't tell you what the owner is trying to build.

So use the numbers to choose where to look, then go and look. A visit, a phone call or ten minutes watching the work will often answer a question that another month of reports can't.

Let the data pick where you look, and let looking tell you why.

What Small Test Would Settle It?

When nobody knows what's possible, prove it small instead of arguing about it. Pick a test you can run quickly, decide beforehand what you'll watch, and see whether the pattern holds.

Writing it down as a hypothesis helps: if we do this, we expect that. It stops the result being read to suit the story you already prefer, which is the quiet risk in every data review.

When Should You Trust Your Judgement Over the Numbers?

Trust your judgement over the numbers when it has been trained by fast, honest feedback in a setting that behaves predictably. Otherwise, go and get better evidence before you overrule the data.

That test comes from Daniel Kahneman and Gary Klein, who represented opposing schools of thought on expert intuition before setting out where they agreed. They concluded that judging an intuition means asking how predictable the environment is and whether the person has had the chance to learn it, and that feeling confident is no guide to being right.

A rep who has worked the same patch for ten years, and seen every hunch confirmed or corrected by the orders that followed, has usually earned that trust; a new rep three months in hasn't, and I was that rep.

When your judgement hasn't earned a say yet, collect the evidence the data can't hold, then decide, say plainly what you weighed beyond the numbers, and write down why. Your intuition improves fastest when you can check it against what happened.

Your judgement earns a say over the numbers when it has a track record of honest feedback behind it.

What Does Data Informed Decision Making Look Like on a Real Decision?

Back to the 240 accounts. Tony didn't stop at the 80/20 split; he helped me dig further into the same data, account by account.

We looked at who was buying which ranges from us, and where someone bought part of a range but not all of it. Then we asked how big each business really was, because a bigger business had far more spending potential than its account showed. And for each one, what didn't I know that I could find out?

Those questions halved the list again. The data still couldn't tell me who would grow, but it had given me a shortlist worth visiting.

One customer stood out for the wrong reasons. He bought only a few bits from us, and what he bought didn't fit the kind of shop he ran.

So I went to see him. When I walked in, there were piles of small boxes with labels on them, waiting by the side of the counter. The shop was incidental: his real business was online, and on the few lines where our pricing was right, he was already winning that business.

We weren't competitive because we hadn't grasped his potential until that visit. Nobody knew what he could sell, so we gave him the margin, through reduced prices, to go after other lines.

We watched one thing: whether the new lines grew the way the first ones had. They did, and his sales and ours grew with them. Within six months he went from my smallest account to my biggest.

The rest followed the same pattern. I kept a close eye on about 15 big customers who gave me half my target, and put my energy into getting a smaller, targeted group to spend. Within three months I went from 75% of target to over 105%, every month.

Run it back through the five steps. The decision was where 140 visits should go, and the sales figures couldn't see potential. The data picked out the accounts worth a closer look, and the visit explained why.

The margin on new lines was the small test, and we knew what we were watching. The judgement that had earned a say was Tony's, not mine: his years in the job told him the 240 deserved a second look.

The spreadsheet ranked him my smallest account; one visit showed he was my biggest opportunity.

Where Does Data Informed Decision Making Go Wrong?

It goes wrong in three predictable ways, and each one looks like good practice from the inside.

The first is using data as cover. The decision is already made and the numbers are gathered afterwards to support it; the tell is a report that arrives with only one possible conclusion in it.

The second is the reverse: calling it judgement when you simply don't like what the data says. If you overrule the numbers, you should be able to name the evidence you're relying on instead.

The third is reaching for more data when the answer needs less. When the reports contradict each other, adding more rarely helps, and deciding which few measures reflect the result you care about usually does.

If you overrule the data, name the evidence you're using instead.

So What Should You Do With Your Data?

Use it to decide where to look, then go and look. That's the whole routine in one line, and it's what got me from 75% of target to over 105%.

The spreadsheet did its job: it showed me where the money was and where it wasn't. What it couldn't do was tell me which small accounts were small for a reason, and that answer only came from asking better questions and walking into a shop.

Next time you sit down with a report, run the five questions before you act. Name the decision, ask what the numbers miss, let the data point you somewhere, test small, and make the call yourself.

Which decision on your list this week is the data quietly making for you?

FAQs

Is data informed decision making better than data driven?

Neither is better in every case. Data-driven works when data is plentiful, fast and clear about what good looks like, such as pricing tests or stock reordering. Data-informed suits decisions where the data is thin or misses what matters, such as hiring or a key account. Most good decisions use both.

How do you know if you're making an objective decision?

You can't remove judgement, but you can make it checkable. Write down the decision and what would change your mind before you look at the data. Agree a small test and what you'll watch, record why you decided, then compare the outcome with what you expected.

What's the difference between data informed and evidence-based decision making?

They overlap heavily. Evidence-based decision making is the name the ISO 9000 quality standards use for deciding on evidence balanced with practical experience, and medicine uses the term too. Data informed decision making focuses on your own data and how far to let it steer the call.

How do schools use data informed decision making?

Schools use it to combine test scores and attendance figures with what teachers see in the classroom. A dip in a reading score shows where to look, and a teacher's knowledge of the pupil explains why. The same steps apply: name the decision, check what the numbers miss and test small.

What should you do when the data and your gut disagree?

Treat the disagreement as a question worth answering. Ask whether your gut has been trained by honest feedback in this kind of decision. If it hasn't, gather the evidence the data can't hold before you overrule it. Our guide to intuitive decision making covers when to trust it.

Written by

Darren Matthews Profile Picture
About
Darren Matthews
After a decade of studying decision-making, I share clear, practical advice to help business professionals make smarter choices.

Your Next Read...

FREE 5-DAY EMAIL COURSE

Catch the Mistake Before You Make
the Decision

I once signed a good deal that nearly cost us payroll. These five emails are the checks that would have caught it, starting with a 30-second test tomorrow.
Free. One email a day, then one every Sunday. Unsubscribe any time.